Shulou(Shulou.com)06/03 Report--

Among the many indicators to evaluate a website, usability is certainly the most important. A website must be available before it can evaluate other aspects. The unavailability of the website can not rely solely on your own people to find or wait for user complaints, but need to be equipped with active means for monitoring. So what we are going to talk to you today is Application Insights, one of the two major services under Azure Monitor. Using Application Insights, we can initiate visits to the website in multiple areas all the time, and immediately alert the relevant personnel to conduct troubleshooting when the usability is poor.

Application Insights, one of the two major services in Azure Monitor, can monitor applications (.net / Nodejs/Java/Python, etc.) running on various platforms hosted in private, hybrid, and any public cloud in near real time. It automatically detects application anomalies and uses its own powerful analysis tools to help diagnose problems, thus improving the performance and availability of the application.

When it comes to this, some friends may ask: what types of surveillance can Application Insights provide for us? Actually, the monitoring types provided by Application Insights are as follows:

Request rate, response time, and failure rate: learn about the most popular pages, periods, and the location of users. See which pages work best. When there are more requests, if the response time is long and the failure rate is high, there may be resource problems. Dependency rate, response time, and failure rate: find out if external services are slowing down. Exception: analyze aggregated statistics, or select a specific instance and drill for stack traces and related requests. Report server and browser exceptions. Page views and load performance: reported by the user's browser. AJAX calls (from web pages): rate, response time, and failure rate. User and session count. Performance counters in Windows or Linux server computers, such as CPU, memory, and network usage. Host diagnostics in Docker or Azure. Diagnostic trace logs in the application: trace events can be associated with requests. Custom events and metrics written by yourself in client or server code to track business events.

Now that you know the type of monitoring, let's take a look at how Aplication Insight monitors the application:

To use Application Insights to monitor the application, we need to create the Application Insights resource in Azure now, and then install a small detection package in the application. The inspection package can detect not only the Web service application, but also the JavaScript in all background components and Web pages, as well as the performance of the host running the application. And send the data to the associated application insights resource and inherit it to the azure monitor

In addition, Application Insights can also set up Web tests that periodically send comprehensive requests to Web services.

Having said that, I'm sure you all have some understanding of Application Insights, so let's take a look at how to use Application Insights to monitor the usability of a website.

Create an Application Insights resource

Click "New Resource"-- "IT& Management tools"-- "Application Insights":

Enter "Resource Group", "name" and "region" and click "review and create":

Click "create" after confirmation:

The creation is complete, as shown in the following figure:

Set up ULR monitoring (website availability monitoring)

In the Application Insights we just created, locate the usability tab and click add Test:

Next, we need to make the following settings:

Test name-this is a test name that we will also use when querying insight data later. Test types-divided into URL tests and multi-step tests. It is important to note here that URL testing is completely free. Multi-step testing can incur costs. To test URL-, we need to monitor the URL test frequency of the website we need to monitor-here I choose the 5-minute test site-because my site is deployed in Southeast Asia, but the users are distributed in multiple regions, so I chose five test sites. Of course, in the actual production environment, we can configure it according to the actual situation. Test success criteria-we can set the test timeout and status code information alerts-here we can set to generate an alert message when a website is detected to be unavailable

Click "context menu"-"Edit Alert" after the creation is completed:

Set the alert rules for on the rule management page, and then click Save:

After a few minutes, we can see the corresponding data under the "Availability" widget on the Overview page:

Click the widget to open a more detailed monitoring dashboard. Selecting details can provide us with more information, such as test duration, available percentage, and number of tests performed. We can also edit or pause the test at the lower right.

Here we can see several more outliers. Hover the mouse over the corresponding outliers to see the duration and location of the test:

After clicking in, you can see detailed monitoring information:

Journal

Because Application Insights and Log Analytics use the same query language, we can click "Log" on the Overview page:

Because "URL test" is stored under the "AvailabilityResults" table, we can use the following command to query:

AvailabilityResults | project-away performanceBucket, operation_Name

We can see the name in the name column, and if we have multiple website usability tests, we can filter accordingly:

Using this monitoring data, we can generate different visual icons or monitor alerts. The scatter chart below, for example, is the average duration by location.

This is how we can use Application Insights to monitor website usability.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.